ABSTRACT
AIM: This study aimed to assess the impact of moderate resistance training on intermuscular adipose tissue (IMAT) in elderly patients with type 2 diabetes and the independent effect of IMAT reduction on metabolic outcomes. METHODS: In this randomized controlled trial, 85 patients with type 2 diabetes were assigned to either the resistance training group (42 participants) or the control group (43 participants) for a 6-month intervention. The primary outcome was changes in IMAT measured by computed tomography scan and magnetic resonance imaging using the interactive decomposition of water and fat with echo asymmetry and least squares qualification sequence. Secondary outcomes included changes in metabolic parameters. RESULTS: Thirty-seven participants in each group completed the study. The IMAT area (measured by a computed tomography scan) in the resistance group decreased from 5.176 ± 1.249 cm2 to 4.660 ± 1.147 cm2, which is a change of -0.512 ± 0.115 cm2, representing a 9.89% decrease from the least-squares adjusted mean at baseline, which was significantly different from that of the control group (a change of 0.587 ± 0.115 cm2, a 10.34% increase). The normal attenuation muscle area (representing normal muscle density) in the resistance group increased from 82.113 ± 8.776 cm2 to 83.054 ± 8.761 cm2, a change of 1.049 ± 0.416 cm2, a 1.3% increase, which was significantly different from that of the control group (a change of -1.113 ± 0.416 cm2, a 1.41% decrease). Homeostasis model assessment 2 of beta cell function (HOMA2-ß; increased from 52.291 ± 24.765 to 56.368 ± 21.630, a change of 4.135 ± 1.910, a 7.91% increase from baseline) and ratio of insulin increase to blood glucose increase at 30 min after the oral glucose tolerance test (∆I30/∆G30; increased from 4.616 ± 1.653 to 5.302 ± 2.264, a change of 0.715 ± 0.262, a 15.49% increase) in the resistance group were significantly improved compared with those in the control group, which had a change of -3.457 ± 1.910, a 6.05% decrease in HOMA2-ß, and a change of -0.195 ± 0.262, a 3.87% decrease in ∆I30/∆G30, respectively. Adjusting for sex, age, diabetes duration, baseline IMAT, and the dependent variable at baseline, linear regression showed that the change in IMAT area was not related to the change in HOMA2 insulin resistance (ß = -0.178, p = .402) or the change in HOMA2-ß (ß = -1.891, p = .197), but was significantly related to the changes in ∆I30/∆G30 (ß = -0.439, p = .047), 2-h postprandial glucose (ß = 1.321, p = .026), diastolic blood pressure (ß = 2.425, p = .018), normal attenuation muscle area (ß = -0.907, p = .019) and 10-year risk of atherosclerotic cardiovascular disease (ß = 0.976, p = .002). CONCLUSION: Low-level, moderate resistance training reduces IMAT content. Even a small reduction in IMAT may be related to a decrease in risk factors for atherosclerotic cardiovascular disease, but this small reduction may not be sufficient to reduce insulin resistance.
ABSTRACT
Background: This study aimed to develop a diabetic kidney disease (DKD) prediction model using long short term memory (LSTM) neural network and evaluate its performance using accuracy, precision, recall, and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Methods: The study identified DKD risk factors through literature review and physician focus group, and collected 7 years of data from 6,040 type 2 diabetes mellitus patients based on the risk factors. Pytorch was used to build the LSTM neural network, with 70% of the data used for training and the other 30% for testing. Three models were established to examine the impact of glycosylated hemoglobin (HbA1c), systolic blood pressure (SBP), and pulse pressure (PP) variabilities on the model's performance. Results: The developed model achieved an accuracy of 83% and an AUC of 0.83. When the risk factor of HbA1c variability, SBP variability, or PP variability was removed one by one, the accuracy of each model was significantly lower than that of the optimal model, with an accuracy of 78% (P<0.001), 79% (P<0.001), and 81% (P<0.001), respectively. The AUC of ROC was also significantly lower for each model, with values of 0.72 (P<0.001), 0.75 (P<0.001), and 0.77 (P<0.05). Conclusion: The developed DKD risk predictive model using LSTM neural networks demonstrated high accuracy and AUC value. When HbA1c, SBP, and PP variabilities were added to the model as featured characteristics, the model's performance was greatly improved.
ABSTRACT
Cancer development is a complex set of proliferative progression, which arises in most cases via multistep pathways associated with various factors, including the tumor microenvironment and extracellular matrix. However, the underlying mechanisms of cancer development remain unclear and this study aimed to explore the role of extracellular matrix in glioma progression. Methods: The expression of type I collagen and fibronectin in tumor tissues from glioma patients was examined by immunofluorescence staining. The correlations between collagen/fibronectin and glioma progression were then analyzed. A 3D collagen/fibronectin cultured system was established for tumor cells culture in vitro. Quantitative, real-time PCR and western blot were used to detect PI3K/ATK and CDC42 signals associated proteins expression in glioma. We used in vitro Cell Counting Kit-8, colony formation, and tumorigenesis assays to investigate the function of PI3K/AKT and CDC42 signals associated proteins. A xenograft glioma mice model was also used to study the anticancer effects of integrin inhibitor in vivo. Results: Our study demonstrated that type I collagen and fibronectin collaborate to regulate glioma cell stemness and tumor growth. In a 3D collagen/fibronectin culture model, glioma cells acquired tumorigenic potential and revealed strengthened proliferative characteristics. More significantly, collagen/fibronectin could facilitate the activation of PI3K/AKT/SOX2 and CDC42/YAP-1/NUPR1/Nestin signaling pathways via integrin αvß3, eliciting sustained tumor growth and cancer relapse. Combination of the integrin signaling pathway inhibitor and the chemotherapeutic agent efficiently suppressed glioma cell proliferation and tumorigenic ability. Conclusion: We demonstrated that type I collagen and fibronectin could collaborate to promote glioma progression through PI3K/AKT/SOX2 and CDC42/YAP-1/NUPR1/Nestin signaling pathways. Blockade of the upstream molecular integrin αvß3 revealed improved outcome in glioma therapy, which provide new insights for eradicating tumors and reducing glioma cancer relapse.
